Reading a Programme in the First Fortnight
People will not tell you the programme is failing. They will tell you, precisely and without hesitation, what they would fix.
Two Weeks to the Truth
When you take over a programme in difficulty, you are handed a version of reality on your first day. It arrives as a pack: the plan, the RAG status, the risk log, the budget position, the milestone tracker. It is tidy, it is confident, and it is almost never true. Not because anyone is lying — but because reported health and real health are different things, and the gap between them is precisely the problem you have been brought in to fix.
The incoming programme manager’s first task is not to act. It is to see. And you have roughly two weeks before you are absorbed into the machine and begin defending the same picture you were sent in to correct. This is a short, practical account of how to use that fortnight — not the full recovery method, which is a longer discipline, but the reading that has to come first.
Reported Health and Real Health
Every troubled programme has two states. The reported state is what the status pack says: on track, amber, recovering. The real state is what is actually true about scope, dependencies, capacity and confidence. In a healthy programme these sit close together. In a troubled one they have drifted apart — and, this is the important part, the drift is itself the strongest signal. A programme reporting green while its people look exhausted and its dependencies sit unresolved is not green with a few issues. It is a programme that has lost the ability to tell itself the truth, and that is a more serious condition than merely being behind.
So the fortnight is not spent checking whether the reported status is accurate. It is spent measuring the distance between the two states — because that distance tells you how much trouble you are really in.
A status report tells you what a programme wants to believe. The first fortnight is for finding out what it actually knows.
What to Look At
Documents lie politely; artefacts tell the truth. In the first two weeks I pay far more attention to the raw material than to the summaries built on top of it.
- The plan at the dependency level, not the milestone level. Milestones are aspirations. Dependencies are physics. Look at what genuinely has to happen before what, and whether each link is agreed by both sides or merely assumed by one.
- The last three status reports side by side. A risk that has been amber for four months is not a risk being managed; it is a problem being tolerated. Watch what quietly moves from ‘in progress’ back to ‘planned’ — that is slippage wearing a disguise.
- The change log. Frequent small scope changes with no corresponding movement in time or cost mean the plan is absorbing pressure silently, which it cannot do forever.
- The actuals against the burn. Where the money and effort are truly going, versus where the plan says they should, is often the fastest route to the real state.
- The decisions that are open. A list of unmade decisions is a more honest risk register than the risk register. Programmes rarely fail on the risks they logged; they fail on the choices no one would make.
Whom to Talk To
The org chart tells you who is responsible. It does not tell you who knows. In the first fortnight the conversations matter more than the documents, and the most valuable ones are rarely at the top.
Talk to the people doing the actual work — the engineers, the analysts, the migration leads. They carry the real state in their heads and are usually relieved to be asked plainly. Talk to the people who own the dependencies you cannot control, especially in other teams and in vendors, because that is where assumed agreements turn out to be one-sided. Talk to whoever has been on the programme longest; they remember the decisions whose consequences you are now living with. And talk to the sceptics — the person everyone describes as difficult is frequently the only one still saying what is true.
The question that opens more than any other is not ‘is it on track’. It is what would you fix first if it were yours? People will not tell you the programme is failing. They will tell you, precisely and without hesitation, what they would fix.
The Signals That Reveal Real Health
Across enough programmes, the signals that separate real health from reported health become recognisable. None of them appears on a RAG dashboard.
- Candour under mild pressure. Ask a straightforward question and watch what happens. In a healthy programme you get a straight answer. In a troubled one you get a caveat, a glance toward a manager, or a story about context. How freely bad news travels upward is the best proxy for real health I know.
- The specificity of confidence. ‘We’ll be fine’ is a worry; ‘we’ll hit it because the last dependency cleared on Tuesday’ is health. Confidence that cannot be traced to a reason is hope in costume.
- The state of the people. Chronic long hours, no one taking leave, a team that has stopped arguing — these tell you more about trajectory than any burndown. Exhausted teams do not recover on schedule.
- What happens to a raised issue. Follow one problem from the person who has it up to the status report. If it shrinks or disappears on the way, you have found your real risk, and it is cultural, not technical.
- Whether anyone can state the goal the same way twice. Ask five people what the programme is for. If you get five different answers, the plan is the least of your problems.
The Discipline of Not Acting Yet
The hardest part of the first fortnight is holding your own hands still. You will be under pressure to demonstrate grip, to make an early call, to reassure the steering committee that the adults have arrived. Resist it. An early intervention based on the reported state simply re-commits you to the fiction. The most valuable thing a new programme manager can do in week one is ask questions with no immediate action attached — and be visibly comfortable not yet knowing the answer.
There is one exception. If in the first days you find something actively unsafe — a go-live about to proceed on data no one trusts, a decision about to be taken on a number known to be wrong — you stop that immediately. Reading the programme does not mean letting it walk off a cliff while you take notes. But such cases are rarer than the pressure to act would suggest, and everything else can wait for the picture to resolve.
By the end of two weeks you should be able to say three things plainly: where the reported and real states diverge most, which unmade decisions are holding everything else hostage, and whom you can trust to tell you the truth from here on. That is not the recovery — the full method comes next. But it is the ground the recovery stands on, and a recovery built on the status pack you were handed on day one will faithfully rebuild the very problem you were sent to solve.
Read first. The fortnight you spend seeing clearly is the cheapest time the programme will ever buy.